Profile banner
Saurjya Sarkar

Saurjya Sarkar

Audio ML engineer and researcher in AI & Music

saurjya
London, United Kingdom
Joined October 2025
Favicon imagesaurjya.github.io
Favicon imagegithub.com
Favicon imageorcid.org
Favicon imagescholar.google.com
Favicon imagex.com
Favicon imageuk.linkedin.com

Network

1.7K connections
🎓
BITS Pilani Alumni & MBA Professionals
🤖
AI/ML & Generative AI Professionals
🔌
Qualcomm & Semiconductor Engineers
🎶
AI & Music/Audio Researchers
🔬
Doctoral & Postdoctoral Researchers
👨‍💻
Senior Technical Professionals
💻
...

Summary

Academic leader in applied audio source separation: Saurjya's PhD and associated papers focus on time-domain models and permutation-invariant training for monotimbral separation (choirs, chamber ensembles and vocal harmonies). He created/used EnsembleSet (a high-quality synthetic dataset) and demonstrated strong cross-dataset gains via pre-training and fine-tuning, work recognised with the Best Student Paper award at WASPAA 2023. ac+2
Bridges research and industry: He has practical industry experience building audio-quality assessment tools at Qualcomm and holds patents related to voice/immersive audio machine learning. In 2025 he transitioned to an industry audio-ML role at Universal Music Group (Abbey Road), applying research expertise to product-scale music/audio engineering problems. github+2
Publishes open code and datasets: maintains public repositories (e.g., EnsembleSep, EnsembleSet-MIDI) and a personal site with demos and code, supporting reproducible research and enabling others to build on his ensemble / harmony separation work. github+2
Practitioner with music-first perspective: musical background (guitar, music club activity) informs an applied approach to research — designing datasets, augmentations and evaluation metrics that reflect musical structure (e.g., harmonic overlap measures) rather than purely engineering benchmarks. ac+1

Work

Education

Projects

Writing

Time-domain Music Source Separation for Choirs and Ensembles

March 1, 2024

PhD thesis presenting time-domain approaches to monotimbral music source separation, introducing EnsembleSet (synthetic dataset), analyses of permutation-invariant training for ensemble/choral separation, and strategies to improve cross-dataset performance.

Favicon imageqmro.qmul.ac.uk

Leveraging synthetic data for improving chamber ensemble separation

October 1, 2023

WASPAA 2023 paper describing a pre-training strategy using synthetic multi-mic EnsembleSet, data augmentation and fine-tuning to improve chamber ensemble separation (Best Student Paper Award at WASPAA 2023).

Favicon imageqmro.qmul.ac.uk

Vocal Harmony Separation using Time-domain Neural Networks

August 1, 2021

Interspeech 2021 paper adapting time-domain encoder-masker-decoder architectures for four-part a cappella vocal separation; includes analysis of harmonic overlap and training strategies for permutation-invariant separation.

Favicon imageisca-archive.org

Hobbies

Guitar playing and active involvement in music performance/production (music club experience, interest in guitar and music creation). ac+1

Audio plugin / VST development and experimentation with audio effects (practical music production and live music tool development). github+1